- This topic has 24 replies, 21 voices, and was last updated 9 years, 11 months ago by .
-
Topic
-
Most people have probably seen reference to the study, some years ago, which suggested that drivers pass closer to cyclists who are wearing a helmet than they do to those who are not. It’s often cited as a reason not to wear a helmet.
Without wanting to make an argument either way as to helmet use, I thought that some people might be interested in a further analysis of the same data (i.e. the researchers got hold of the raw data from the original experiment and looked at it again to see whether it was properly interpreted). This study went into rather more depth, comparing a number of factors and considering their statistical significance rather than just drawing simple numbers.
Short version of the conclusions:
1. Yes, cars pass closer on average if you are wearing a helmet.
2. That only applies to cars passing more than 2m away. Cars that give you 2m+ clearance will give an average 7cm more room to somebody without a helmet (2.27m v 2.20m).
3. The overwhelming majority of passes took place 1-2m away. They gave, on average, 1-2cm more clearance to those without a helmet (and this figure was not statistically significant). Also not statistically significant were the figures for 0.5-0.75m passes (no difference) and less than 0.75cm (those wearing helmets were given 5cm MORE room – 66cm v 61cm).
4. Overall, since the only statistically significant increase is in passes at a good distance “these results do not support the idea that any substantive risk reduction can be gained from not wearing a helmet”.
5. The most significant correlations were actually road type (drivers on the “regular urban streets” of Bristol gave less clearance than those on the “residential” and “rural” roads of Salisbury) and distance to kerb (the further the cyclist was from the kerb, the closer the cars passed).
As I said, not trying to argue one way or the other, but this study is often brought up and i thought it might be useful to see that it does not in fact support the proposition for which it is generally cited. The full text of the reanalysis is here: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3783373/
- You must be logged in to reply to this topic.